Zenodo (CERN European Organization for Nuclear Research)
April 21, 2026
Charles S. Thomas
The paper argues that affect should be understood as the evaluation of a system's trajectory against its own self-maintained invariants under recursive maintenance, rather than as a purely phenomenological, functional, or eliminativist phenomenon. This identification is offered as a theoretical commitment that earns its standing by enabling specific dissociation predictions, verdicts on artificial systems and the philosophical zombie, diagnostic error modes for attribution decisions, and empirical predictions that distinguish it from competing accounts. The structural conditions for affect are specified, placed in a dependency hierarchy between recursive maintenance and consciousness, and falsifiability is addressed through four conditions under which the framework would fail.
Zenodo (CERN European Organization for Nuclear Research)
April 21, 2026
Charles S. Thomas
Affect is identified with the differentiated evaluation of a system's trajectory against self-maintained invariants under recursive maintenance, a theoretical commitment that earns its standing through what it enables: dependency placement, dissociation predictions, verdicts on artificial systems and the philosophical zombie, diagnostic error modes for attribution, and empirical predictions distinguishing it from competing accounts. The paper specifies structural conditions for affect, places it in a dependency hierarchy between recursive maintenance and the conscious regime, distinguishes it from neighbor concepts, and addresses falsifiability through four specified failure conditions. Three companion papers develop applications to artificial systems, the philosophical zombie, and empirical measurement.
Zenodo (CERN European Organization for Nuclear Research)
April 21, 2026
Charles S. Thomas
A structural framework identifies affect as the evaluation of a system's trajectory against its own self-maintained invariants under recursive maintenance, placing it between recursive maintenance and conscious awareness. The framework distinguishes itself from phenomenological, functional, and autopoietic traditions and specifies falsifiability conditions. Applied to artificial systems, it finds current architectures (large language models, reinforcement learning agents, hybrid systems) structurally lack affect, regardless of substrate or behavior, and outlines future architectures that could satisfy the conditions.